We propose to establish a Clinical-Epidemiology Center Consortium consisting of the collaboration of the following Institutions and their respective consortium partners: University of Michigan Medical Center- University of Michigan Cancer Center, Dartmouth Medical School- Norris Cotton Cancer Center, the Dana Farber Cancer Institute, Wellesley-St. Michael Hospital of Toronto, and the Henry Ford Medical System. We propose to pool our clinical experience to address the needs of the Early Detection Network by providing a large human subject pool for prevention clinical research, by devising innovative approaches to the validation of biomarkers, by sharing samples with Network laboratories and clinical centers, and by closely interacting with Network Biomarkers Development and Validation Laboratories. The following Specific Aims are proposed: l. Initiate validation studies for selected biomarkers for colorectal adenomas and cancers; 2. Refine statistical criteria for a clinical biomarker validation program, 3. Establish organizational structure for clinical studies of biomarkers for research in colon and other cancers, 4. Establish mechanisms for use of development funds; and 5. Describe quality assurance procedures. Specific Aim #1 will be addressed through a four stage biomarker validation paradigm. The first stage consists of characterization studies to individual variability of a biomarker. The second stage consists of an intermediate size cross sectional validation trial. The third stage consists of a longitudinal observational trial. The fourth stage consists of a randomized, double blinded chemoprevention validation trial. In this application, we propose to validate five potential serum biomarkers and aberrant crypt foci as a pathologic biomarker. Specific Aim #2 will be addressed through the development and prospective testing of newly developed statistical methods designed to assess biomarker efficacy in panels. Specific Aim #3 will be addressed by establishing an Operations Center at the University of Michigan, a Data Management Unit at Dartmouth Medical School, and staffing all clinical sites with a dedicated clinical research associate experienced in sample collection, handling, tracking, shipment and data management. Specific Aim #4 will be addressed through two proposed development projects, one assessing K-ras in stool guaiac samples and a second studying cytochrome P-450 isozymes in tissue samples. The a procedure for review of development projects is described. Specific Aim #5 is addressed through description of data management, laboratory, pathology, and endoscopic quality control procedures. Our consortium, while focused upon the validation of biomarkers for colorectal cancer provides extensive, diverse human populations from which to validate early detection biomarkers for high priority cancers.